"Thomas E. Dickey" <[EMAIL PROTECTED]> writes:
> changes were made to autoconf
What changes are you talking about, exactly?
> which made it incompatible with known, widely-used versions of
> Bourne shell
Which versions of the Bourne shell are you talking about here, exactly?
And which softare i
Eric M. Monsler writes:
> How can I get the resulting Makefile to end up with the default
> transform line of:
>
> transform=s,$$,$(VERSION),
>
> in the absence of any overriding or additional configure options?
Check how Automake handles versioned installs in its top-level
Makefile.am.
--
Pete
Raphaël Poss writes:
> This does not:
> ---8<---
> AC_INIT(tools, 0.1)
> AM_INIT_AUTOMAKE
> AC_OUTPUT
> ---8<---
AM_INIT_AUTOMAKE calls AC_PROG_INSTALL which needs install-sh.
> This does not, either:
> ---8<---
> AC_INIT(tools, 0.1)
> AC_CANONICAL_HOST
> AC_OUTPUT
> ---8<---
AC_CANONICAL_HOST
Fred Cook writes:
> #include FOO
>
> int main()
> {
>return 0;
> }
>
> Where FOO is a #define that will be set by a previous macro.
Use a shell variable.
--
Peter Eisentraut [EMAIL PROTECTED]
zaufi writes:
> and autoheader fail on this with the following error:
> autoheader: missing template: TEST
> Use AC_DEFINE([TEST], [], [Description])
>
> but manual says that AC_DEFINE with 1 argument is valid call (and I think so)
It's valid if and only if you don't use autoheader.
--
Peter Ei
Hi all!
I found this strange behaviour when try to make libmcrypt support in my test
project.
I make a test config.in like this:
AC_INIT([test],[1.0])
AC_CONFIG_HEADER([config.h])
AC_DEFINE(TEST)
AC_OUTPUT
and autoheader fail on this with the following error:
autoheader: missing template: TEST
U
On Fri, 21 Feb 2003, John Burger wrote:
> It seems to be explicitly against the philosophy of Autoconf to do
> anything in response to particular hardware or OS versions. Rather,
back to the original point - changes were made to autoconf which made
it incompatible with known, widely-used version
